Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The waiting room seemed to shrink, the floral wallpaper closing in on her. The nurse finally called, "Mrs. Chen?" and Emily practically flew out of her seat. This was it – the culmination of weeks of relentless digging.

"Have a seat," the physician said, motioning towards a chair, a slightly bewildered expression on his face. She plopped down, unable to keep still. Her foot tapped a frantic rhythm against the floor.

"We have a slight issue," the doctor began, adjusting his glasses. Emily held her breath. "It seems your records… they've been inadvertently cross-referenced with… Mr. Peterson's."

She couldn't help the ecstatic shout that escaped her. She clapped her hands together, a giddy wave of relief washing over her.

"Brilliant," she breathed, "Just absolutely brilliant."

Emotion: excited

Cluster: Positive / Joy
PC1 (Valence): 3.21 Positive
PC2 (Disposition): -0.05

Role in Research

This story is one of 1,000 stories generated for the emotion excited. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 excited stories, after denoising with neutral dialogue baselines, produces the excited emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the excited vector is projected through the unembedding matrix.

Promoted:
de0.809
!0.564
excitedly0.500
🤩0.491
!"0.466
Suppressed:
S-0.721
ness-0.519
😞-0.424
grieve-0.310
😔-0.300